diff --git a/ggml/src/ggml-sycl/fattn-tile.hpp b/ggml/src/ggml-sycl/fattn-tile.hpp index c4d24613a5..b4d4e0ae90 100644 --- a/ggml/src/ggml-sycl/fattn-tile.hpp +++ b/ggml/src/ggml-sycl/fattn-tile.hpp @@ -1252,6 +1252,16 @@ static void launch_fattn_tile_switch_ncols1(ggml_backend_sycl_context & ctx, ggm return; } + { + constexpr int cols_per_block = ncols2*2; + const int nwarps = ggml_sycl_fattn_tile_get_nthreads (DKQ, DV, cols_per_block, cc) / warp_size; + const int nbatch_fa = ggml_sycl_fattn_tile_get_nbatch_fa(DKQ, DV, cols_per_block, cc); + launch_fattn, warp_size> + (ctx, dst, nwarps, nbytes_shared, nbatch_fa, true, true, false); + return; + } + GGML_ABORT("fatal error"); }